Background: Chronic kidney disease is one of the leading causes of mortality in patients with sickle cell disease. However, it has been almost exclusively studied in patients with the SS phenotype and in high-income countries, despite more than 80% of patients living in Africa. We looked for the determinants of glomerulopathy in a multinational cohort of patients with sickle cell disease of diff erent phenotypes in sub-Saharan Africa. Methods: In the CADRE cohort, we prospectively included patients 3 years and older with sickle cell disease of all haemoglobin phenotypes in Cameroon, Côte d'Ivoire, Mali, and Senegal. All individuals were assessed at steady state. The main outcome of interest was albuminuria defi ned as a urine albumin-to-creatinine ratio of greater than 30 mg/g. We investigated the clinical and biological determinants (including haemolysis markers) of albuminuria in two main phenotype groups (SS and Sβ° SC and Sβ+) with further stratifi cation by age and country. Findings: The study is ongoing because of follow-up. 2582 patients with sickle cell disease were included (1776 SS, 136 Sβ°, 511 SC, and 159 Sβ+). 644 patients with the SS and Sβ° phenotypes (33·7%, 95% CI 31·6-35·8) and 110 with the SC and Sβ+ phenotypes (16·4%, 13·6-19·2) had albuminuria. In the SS and Sβ° group, albuminuria was detected in 144 (27%) of 527 children younger than 10 years and its frequency increased with age (29 [48%] of 60 patients aged >40 years). Multivariable analysis showed that albuminuria was associated with age (odds ratio 1·43, 95% CI 1·20-1·71; p<0·0001), female sex (1·35, 1·02-1·82; p=0·045), low haemoglobin (0·79, 0·66-0·93; p=0·006), high lactate dehydrogenase concentrations (1·33, 1·14-1·58; p=0·0009), and, using Côte d'Ivoire as the reference, Mali (2·49, 1·64-3·79; p=0·042) and Cameroon (1·59, 1·01-2·51; p=0·0007) in patients with the SS and Sβ° phenotypes. The magnitude of the association of albuminuria with haemoglobin and lactate dehydrogenase concentrations increased with age. In the SC and Sβ+ patients, only low haemoglobin (0·69, 0·48-0·97; p=0·029), high blood pressure (1·63, 1·17-2·27; p=0·0017), and Mali (3·75, 1·75-8·04; p<0·0001) were associated with albuminuria. Interpretation: Hyperhaemolysis is associated with albuminuria, with an age-dependent eff ect, in the SS and Sβ° phenotypes only, suggesting a diff erent pathological mechanism for glomerular disease in the patients with SC andSβ+ phenotypes. However, both phenotypes are associated with a high prevalence of albuminuria in childhood. Therefore, screening for albuminuria is advised in African children with sickle cell disease to detect early renal damage.
